Package index • sampleSizeCalc

Unified Three-Axis API / 统一三轴入口

Primary entry points supporting 39+ study design combinations (design x task x outcome) and legacy single-keyword compatibility.

calculate(): Unified sample size calculator (three-axis API)
sample_size_calculator() sampleSizeCalculator(): Sample size calculator for clinical study designs (legacy API)

Results Display & Visualisation / 结果输出与绘图

Default print and summary reports include plain-text formulas, LaTeX source, calculation steps, and references. Plot methods provide visual diagnostics.

print(<sample_size_result>): Print method for sample_size_result objects
summary(<sample_size_result>): Summary method for sample_size_result objects
plot(<sample_size_result>): Plot sample size calculation result
plot_power_curve(): Plot power curve
plot_sample_size_comparison(): Plot sample size comparison

Clinical Trials (RCTs) / 随机对照试验

Calculations for parallel-group randomised controlled trials across continuous, binary, and time-to-event outcomes, covering superiority, non-inferiority, and equivalence.

calculate_two_means(): Two independent means comparison calculation
calculate_two_proportions(): Two independent proportions comparison calculation
calculate_survival(): Survival analysis (log-rank test) calculation
calculate_superiority_means(): Superiority test (means) calculation
calculate_superiority_proportions(): Superiority test (proportions) calculation
calculate_non_inferiority_means(): Non-inferiority test (means) calculation
calculate_non_inferiority_proportions(): RCT non-inferiority for proportions (fills the legacy margin gap)
calculate_equivalence_means(): Equivalence test (means) calculation
calculate_equivalence_proportions(): Equivalence test (proportions) calculation

Paired, Single-Arm & Crossover Designs / 配对、单臂与交叉设计

Within-subject paired comparisons, single-arm trials against historical benchmarks, Simon two-stage designs, A’Hern single-stage, and 2x2 bioequivalence crossover trials.

calculate_paired_means(): Paired means comparison calculation
calculate_paired_proportions(): Paired binary outcomes (McNemar test sample size)
calculate_single_arm_mean(): Single-arm trial: continuous outcome vs a fixed reference value n = (Z_a/2 + Z_b)^2 * sigma^2 / delta^2 (Machin Ch.5, single-sample)
calculate_single_arm_proportion(): Single-arm trial: binary outcome vs a fixed reference proportion
calculate_simon_two_stage(): Single-arm phase II: Simon two-stage design (exact search)
calculate_ahern(): Single-arm phase II: A'Hern single-stage design (exact search)
calculate_crossover_mean(): 2x2 crossover, continuous outcome (within-subject SD parameterization)
calculate_bioequivalence(): Bioequivalence (2x2 crossover, exact TOST noncentral t, 80-125%)

Observational & Epidemiological Studies / 观察性流行病学研究

Sample size calculations for prospective cohort studies (cumulative incidence and person-time rate), unmatched/matched case-control studies, and cross-sectional prevalence surveys.

calculate_cohort_proportion(): Cohort study: exposed vs unexposed cumulative incidence (1:m allocation)
calculate_cohort_rate(): Cohort study: incidence rate comparison (person-time, Poisson)
calculate_case_control(): Case-control study calculation
calculate_case_control_matched(): Individually matched (1:1) case-control study
calculate_crosstab_association(): Cross-sectional study: association between exposure and prevalence

Single-Sample Precision Estimation / 单组参数置信区间估计

Determination of sample sizes required to estimate a population proportion (prevalence), continuous mean, or Pearson correlation coefficient with a target margin of error.

calculate_single_proportion(): Internal calculation: single proportion precision (absolute or relative)
calculate_single_mean(): Internal calculation: single mean precision
calculate_single_correlation(): Internal calculation: single correlation coefficient precision (Fisher z / Machin)

Diagnostic Accuracy Studies / 诊断试验准确性研究

End-to-end sample size determination for diagnostic accuracy studies: sensitivity/specificity estimation, single/dual-endpoint hypothesis testing, paired/unpaired comparisons, ROC-AUC (estimation, testing, margins), and likelihood ratios.

calculate_diagnostic_accuracy_v2(): Upgraded diagnostic accuracy: Zhou Eq 6.1 (CI width + power), Eq 6.2 (precision), Eq 6.3 (prevalence guarantee) — Zhou 2011 pp.194-196
calculate_diagnostic_superiority_single(): Diagnostic superiority (single test) calculation
calculate_diagnostic_comparison_two(): Diagnostic comparison (two tests) calculation
calculate_diag_compare_margin_prop(): Two-proportion margin designs (non-inferiority / equivalence)
calculate_auc_single_v2(): Upgraded AUC estimation: three variance functions, user-set case:control ratio R — Zhou 2011 pp.196-199
calculate_auc_superiority(): AUC superiority test calculation
calculate_auc_comparison(): AUC comparison (two tests) calculation
calculate_diag_auc_compare_margin(): AUC margin designs (non-inferiority / equivalence of two ROC areas)
calculate_likelihood_ratio_v2(): Likelihood ratio sample size via log(LR) CI (Hajian-Tilaki Eq 6.10-6.14)
screen_required(): Total screening size with probability guarantee (Zhou Eq 6.3)
mrmc_power(): MRMC study power (Obuchowski-Rockette mixed model, paired-patient paired-reader design)

Complex & Modern Clinical Designs / 复杂与现代临床研究设计

Methodologies for cluster randomised trials (design effect, Machin Eq 12.10 with the Student t correction, optional variable cluster size), stepped-wedge cluster trials, ordered categorical outcomes, clinical prediction model sample size, and interrupted time series.

calculate_cluster_mean(): Cluster RCT wrapping the two-means kernel
calculate_cluster_proportion(): Cluster RCT wrapping the two-proportions kernel
cluster_design_from_indiv(): Cluster RCT with Machin Eq 12.10 (DE + Student t correction)
cluster_size_inflation(): Design effect for a cluster randomised trial (Machin Eq 12.5 / 12.7)
apply_cluster_design(): Apply the cluster design effect to an individually-randomised result
calculate_stepped_wedge(): Internal calculation: stepped wedge power (Hussey & Hughes 2007 Eq 8)
calculate_two_ordinal(): Two-group comparison of an ordered categorical outcome (Whitehead)
calculate_predictive_riley_binary(): Internal calculation: prediction model development (Riley 2020 BMJ method)
calculate_predictive_timetoevent(): Internal calculation: prediction model for a time-to-event outcome (Riley 2020 criterion T1: person-years for precise cumulative incidence) Anchor: lambda=0.1/year, t=10, delta=0.05 -> T >= 2366 person-years.
calculate_predictive_epv(): Predictive/prognostic model development: events-per-variable rule
calculate_its_level(): Interrupted time series: level-change power (AR(1)-adjusted, experimental)
calculate_kappa(): Internal calculation: Cohen's kappa sample size (Cantor 1996)

Internal Validation & Utilities / 内部校验与辅助工具

Lower-level argument validation, standard normal quantiles, and assertion utilities.

validate_sample_size_result(): Validate a sample_size_result object
assert_sample_size_result(): Assert an object is a valid sample_size_result
validate_design_args(): Validate design-specific arguments
validate_alpha(): Validate alpha (type I error) parameter
validate_power(): Validate power parameter
validate_alternative(): Validate the alternative argument
new_sample_size_result(): Internal constructor for sample_size_result objects